Implementasi Model Yolov8 untuk Deteksi Jenis Sampah Organik dan Anorganik Berbasis Android

The mismanagement of waste poses serious environmental and public health issues in Indonesia, exacerbated by the increasing volume of waste due to population growth. To address this problem, this research develops a mobile application based on Flutter, utilizing YOLOv8 object detection technology to classify organic and inorganic waste. The application aims to simplify household waste sorting, raise public awareness, and support better and more sustainable waste management. The research methodology involves using a dataset of waste images trained with the YOLOv8 algorithm via google colab. The dataset is divided into training (70%), testing (20%), and validation (10%) portions. The model training process is conducted over 25 and 50 epochs, showing improved accuracy with more epochs. At the 50th epoch, the model achieved a precision of 0.81 and a recall of 0.61, demonstrating good performance in detecting and classifying waste. The implementation of this application is expected to facilitate waste sorting, reduce environmental pollution, and improve public health. Recommendations for further development include enhancing detection accuracy, expanding the range of detectable waste types, and optimizing application performance to ensure a better user experience.